Abstract
<title>Abstract</title> <p> This paper examines the mechanism through which cooperative procurement guarantees generate persistent sugarcane concentration in Maharashtra, India. Using a balanced district-year panel of 238 observations across 34 agricultural districts from 2016–17 to 2022–23, four estimation methods are applied: panel fixed effects with continuous mill density (dose-response), a coordination amplification interaction specification, panel quantile regression (Canay, 2011), and a two-estimator causal analysis combining two-way fixed effects DiD and the Generalized Synthetic Control (Xu, 2017). Cooperative mill density has a monotonic dose-response relationship with sugarcane acreage share: each doubling of the mill count adds 3.2 percentage points (log_mills = 0.047, SE = 0.016, p = 0.005). Mill density amplifies year-on-year persistence continuously from 34.4 per cent at zero mills to 101.1 per cent at 19 mills (interaction = 0.223, SE = 0.047, p < 0.001). Panel quantile regression reveals that persistence exceeds unity at the 90th percentile of the district sugarcane distribution, indicating self-amplifying coordination dynamics in the most institutionally embedded districts. AgriStack Farmers’ Registry activation produced no detectable positive reallocation (TWFE ATT = − 0.037, p = 0.088; GSC ATT = − 0.005), confirming that digital information provision cannot dissolve a coordination equilibrium. Smallholder-intensive districts show greater sugarcane concentration when mill density amplifies persistence (SmallFarmer = 1.738, p = 0.039). The results are consistent with a stag hunt coordination game in which cooperative mill viability depends on aggregate cane supply, making individual deviation individually irrational regardless of alternative crop price signals. <bold>JEL Classification</bold> : Q12, Q18, O13, C23, C31, C35, D02 </p>